Epilepsy affects nearly 3 million Americans and approximately 30% of these patients continue to have seizures despite treatment with multiple antiepileptic drugs. Improved understanding of the cellular and molecular changes underlying seizures, and antiepileptic drug action, is essential to advance the care of veterans suffering from medically refractory epilepsy. Among the changes seen in epilepsy is a reduction of neuronal membrane transporters for the inhibitory neurotransmitter 3-aminobutyric acid (GABA) [GABA transporter type 1 (GAT1)] (During et al., 1995; Patrylo et al., 2001). Classically, GABA transporters were thought to function by removing GABA from the extracellular space, thereby terminating GABA actions. It is now clear that GABA transporters can also operate in the reverse direction to release GABA into the extracellular space, a process termed nonvesicular release. Nonvesicular GABA release is augmented by neuronal depolarization and alterations of ionic gradients, conditions that exist during epileptic seizures. Nonvesicular GABA release may be particularly important during prolonged seizures when vesicular GABA release at synapses is compromised by reductions in extracellular [Ca2+], GABA packaging into synaptic vesicles is limited by energy failure, and GABA synthesis is stimulated by low [ATP], causing an increase in cytoplasmic [GABA] that favors nonvesicular release (Erecinska et al., 1984; Hablitz and Heinemann, 1987; Allen et al., 2004). This proposal will test the hypothesis that nonvesicular GABA release has an important antiepileptic action and GAT1 downregulation in epilepsy is maladaptive. Specifically, it is hypothesized that the loss of nonvesicular GABA release mediated by GAT1 will exacerbate seizures despite elevated basal concentrations of ambient GABA. Consistent with this, studies on animals with genetic deletion of GABA transporter type 1 [GAT1 knockout (GAT1 KO)] have identified a heightened sensitivity to chemoconvulsants (Chiu et al., 2005). GAT1 function in epilepsy will be evaluated by three specific aims: Specific Aim 1 will test the hypothesis that nonvesicular GABA release in brain slices from GAT1 KO animals is reduced compared to wild type littermates. Specific Aim 2 will test the hypothesis that AEDs that increase GABA concentration or GAT1 surface expression will enhance nonvesicular release from wild type but not GAT1 KO neurons. Specific Aim 3 will test the hypothesis that nonvesicular release mediated by GAT1 affects seizure threshold in vivo and limits seizure frequency/severity in epileptic mice. Identification of changes in nonvesicular release in GAT1 knockout neurons and determination of AED effects on nonvesicular release in Specific Aims 1-2 will provide a basis for interpretation of in vivo data.